CN111753633B - Target detection method and device - Google Patents

Abstract

The application provides a target detection method and a target detection device, wherein the method comprises the following steps: detecting a target video from a video stream acquired by a camera when a target object executes a specified operation, wherein the target video is a video containing an operation to be checked executed by the target object in the process of executing the specified operation; determining the assessment score of the target object for executing the operation to be examined according to the target video; and determining the assessment score of the target object for executing the specified operation according to the assessment score of the target object for executing the operation to be examined. By the method, the process of executing the specified operation on the target object can be automatically checked, and meanwhile, the objectivity and the accuracy of the checking results on different target objects can be effectively ensured because unified checking standards are adopted for different target objects.

Description

Target detection method and device

technical field

The present application relates to the technical field of video processing, and in particular to a target detection method and device.

Background technique

In some applications, humans are often used to monitor the process of the target object performing a specified operation, and to evaluate the specified operation performed by the target object.

Taking the chemical experiment operation performed by the candidates as an example, one or more invigilators are responsible for monitoring the process of each candidate performing the chemical experiment operation in the examination room, and determining the examination score for the candidate's chemical experiment operation. Such processing not only consumes labor costs, but also has different assessment standards for the same designated operation by different invigilators, resulting in inaccurate assessment scores for the chemical experiment performed by the invigilators.

Contents of the invention

In view of this, the present application provides a target detection method and device.

According to a first aspect of an embodiment of the present application, a target detection method is provided, the method comprising:

Detecting a target video from the video stream collected by the camera when the target object performs a specified operation, the target video is a video containing the operation to be assessed performed by the target object during the execution of the specified operation;

determining the assessment score for the target object to perform the operation to be assessed according to the target video;

The assessment score for the target object to perform the specified operation is determined according to the assessment score for the target object to perform the operation to be assessed.

In one embodiment, the detecting the target video from the video stream collected by the camera when the target object performs a specified operation includes:

Identifying the video frames in the video stream in sequence, and determining the target when it is identified that there is a first event in the first video frame of the video stream indicating that the target object starts to perform the operation to be assessed The object starts to perform the operation to be assessed, and when it is recognized that a second event indicating that the target object ends the operation to be assessed exists in a second video frame after the first video frame, determining the target object End the execution of the operation to be assessed;

Determining a section of video in which the first video frame is the first video frame and the second video frame is the last video frame in the video stream as the target video corresponding to the operation to be assessed.

In an embodiment, the determining the assessment score of the target object performing the operation to be assessed according to the target video includes:

Inputting the target video into a trained deep learning model to obtain assessment information corresponding to the operation to be assessed;

An assessment score of the operation to be assessed is determined according to the assessment information.

In one embodiment, the method further includes:

determining the assessment score of the operation result information obtained by the target object after performing the specified operation;

The determining the assessment score of the target object performing the specified operation according to the assessment score of the target object performing the operation to be assessed includes:

The assessment score of the target object performing the specified operation is determined according to the assessment score of the operation result information and the assessment score of the target object performing the operation to be assessed.

In one embodiment, the specified operation performed by the target object is: the examinee performs the experiment of measuring the pH value of the solution;

The determination of the assessment score of the operation result information obtained by the target object after performing the specified operation includes:

Detect the discoloration area on the pH test paper and the pH value of the solution measured by the examinee from the video stream;

Using the color information of the discolored area as a key to find the corresponding pH value in the color comparison card;

The assessment score is determined based on the found PH value and the solution PH value measured by the examinee.

In one embodiment, the specified operation performed by the target object is: the examinee performs the experiment of measuring the pH value of the solution;

The operation to be assessed is: the examinee performs the liquid taking operation and the pH value reading colorimetric operation in the experimental process of measuring the pH value of the solution;

The liquid-taking operation refers to the operation of dipping the liquid to be tested with a glass rod, and dropping the liquid to be tested on the pH test paper placed on the drip plate;

The pH value reading operation refers to the operation of using a color comparison card and a color-changing pH test paper to perform color comparison.

According to a second aspect of the embodiments of the present application, there is provided a target detection device, the device comprising:

The detection module is configured to detect the target video from the video stream collected by the camera when the target object performs a specified operation, and the target video includes the operation to be assessed performed by the target object during the specified operation video of

A first assessment module, configured to determine the assessment score for the target object to perform the operation to be assessed according to the target video;

The second assessment module is configured to determine the assessment score for the target object to perform the specified operation according to the assessment score for the target object to perform the operation to be assessed.

In one embodiment, the detection module detects the target video from the video stream collected by the camera when the target object performs a specified operation, including:

Identifying the video frames in the video stream in sequence, and determining the target when it is identified that there is a first event in the first video frame of the video stream indicating that the target object starts to perform the operation to be assessed The object starts to perform the operation to be assessed, and when it is recognized that a second event indicating that the target object ends the operation to be assessed exists in a second video frame after the first video frame, determining the target object End the execution of the operation to be assessed;

Determining a section of video in which the first video frame is the first video frame and the second video frame is the last video frame in the video stream as the target video corresponding to the operation to be assessed.

In an embodiment, the first assessment module determines the assessment score of the target object performing the operation to be assessed according to the target video, including:

Inputting the target video into a trained deep learning model to obtain assessment information corresponding to the operation to be assessed;

An assessment score of the operation to be assessed is determined according to the assessment information.

In one embodiment, the device further includes:

The third evaluation module is used to determine the evaluation score of the operation result information obtained by the target object after performing the specified operation;

The second assessment module determining the assessment score of the target object performing the specified operation according to the assessment score of the target object performing the operation to be assessed includes:

The assessment score of the target object performing the specified operation is determined according to the assessment score of the operation result information and the assessment score of the target object performing the operation to be assessed.

In one embodiment, the specified operation performed by the target object is: the examinee performs the experiment of measuring the pH value of the solution;

The third evaluation module determines the evaluation score of the operation result information obtained by the target object after performing the specified operation includes:

Detect the discoloration area on the pH test paper and the pH value of the solution measured by the examinee from the video stream;

Using the color information of the discolored area as a key to find the corresponding pH value in the color comparison card;

The assessment score is determined based on the found PH value and the solution PH value measured by the examinee.

In one embodiment, the specified operation performed by the target object is: the examinee performs the experiment of measuring the pH value of the solution;

The operation to be assessed is: the examinee performs the liquid taking operation and the pH value reading colorimetric operation in the experimental process of measuring the pH value of the solution;

The liquid-taking operation refers to the operation of dipping the liquid to be tested with a glass rod, and dropping the liquid to be tested on the pH test paper placed on the drip plate;

The pH value reading operation refers to the operation of using a color comparison card and a color-changing pH test paper to perform color comparison.

According to a third aspect of the embodiments of the present application, an electronic device is provided, and the device includes a readable storage medium and a processor;

Wherein, the readable storage medium is used for storing machine-executable instructions;

The processor is configured to read the machine-executable instructions on the readable storage medium, and execute the instructions to implement the steps of the object detection method provided in the embodiments of the present application.

According to a fourth aspect of the embodiments of the present application, a computer-readable storage medium is provided, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the target detection provided by the embodiments of the present application is realized method steps.

Applying the embodiment of this application, by detecting the target video from the video stream collected by the camera when the target object performs a specified operation, determine the assessment score of the target object performing the operation to be assessed according to the target video, and determine the assessment score of the target object performing the operation to be assessed according to the target object. The assessment score determines the assessment score for the target object to perform the specified operation, which realizes the automatic assessment of the process of the target object performing the specified operation. At the same time, because the uniform assessment standard is adopted for different target objects, it can effectively ensure the assessment of different target objects Objectivity and accuracy of results.

Description of drawings

FIG. 1 is a schematic diagram of an application scenario provided by an exemplary embodiment of the present application;

FIG. 2 is a flow chart of an embodiment of a target detection method provided by an exemplary embodiment of the present application;

FIG. 3 is an implementation process of step 202 provided by an exemplary embodiment of the present application;

FIG. 4 is an implementation process after step 203 provided by an exemplary embodiment of the present application;

Fig. 5 is a block diagram of an embodiment of a target detection device provided by an exemplary embodiment of the present application;

Fig. 6 shows a hardware structure diagram of an electronic device according to an exemplary embodiment of the present application.

Detailed ways

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numerals in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with aspects of the present application as recited in the appended claims.

The terminology used in this application is for the purpose of describing particular embodiments only, and is not intended to limit the application. As used in this application and the appended claims, the singular forms "a", "the", and "the" are intended to include the plural forms as well, unless the context clearly dictates otherwise. It should also be understood that the term "and/or" as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

It should be understood that although the terms first, second, third, etc. may be used in this application to describe various information, the information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of the present application, first information may also be called second information, and similarly, second information may also be called first information. Depending on the context, the word "if" as used herein may be interpreted as "at" or "when" or "in response to a determination."

In order to facilitate the understanding of the present application, firstly, the application scenarios of the object detection method proposed in the present application are described as examples.

In an exemplary application scenario, a camera can be used to capture the video stream of candidates performing chemical experiment operations, such as the test of measuring the pH value of a solution, and determine the test scores for candidates performing chemical experiment operations based on the target detection method proposed in this application. Based on this, in the application scenario illustrated in FIG. 1 , it may include: an examinee 101 , an operating console 102 , a camera 103 , a network 104 and an electronic device 105 .

Wherein, the network 104 is used to provide a communication link between the camera 103 and the electronic device 105, and may include a wireless link, a wired link, or an optical fiber cable, etc., which is not limited in this application.

Experimental equipment (not shown in FIG. 1 ), such as glass rods, measuring cups, test tubes, etc., required for performing chemical experiment operations can be placed on the operating table 102 for candidates to perform chemical experiment operations.

The camera 103 may be implemented as a bullet camera, a fisheye camera, a dome camera, etc. The present application does not limit the specific implementation form of the camera 103 . In the embodiment of the present application, the camera 103 may face the operation table 102 to collect video streams of candidates performing chemical experiment operations. In a preferred embodiment, in order to ensure the accuracy of the examination results, it is set that one camera only corresponds to one operation platform, that is, one camera is only responsible for collecting the video stream of one candidate performing chemical experiment operations.

The electronic device 105 may be a hardware device that supports network connection to provide various network services, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, servers and so on. In this embodiment of the application, the electronic device 105 can obtain the video stream collected by the camera 103 when the candidate performs the chemical experiment operation, and execute the target detection method provided by the application based on the video stream, so as to automatically determine that the candidate performs the chemical experiment operation assessment score.

Based on the application scenario illustrated in Figure 1, the object detection method provided by this application will be described through specific embodiments:

Please refer to FIG. 2 , which is a flowchart of an embodiment of a target detection method provided by an exemplary embodiment of the present application. In an example, the method may be applied to an electronic device, such as the electronic device 105 illustrated in FIG. 1 .

As shown in Figure 2, it includes the following steps:

Step 201: Detect the target video from the video stream collected by the camera when the target object performs a specified operation, the target video is a video containing the operation to be assessed performed by the target object during the specified operation.

As an example, the electronic device may obtain in real time the video stream collected by the camera when the target object performs a specified operation.

As another example, after the target object completes the specified operation, the electronic device may also acquire the video stream collected by the camera when the target object performs the specified operation.

In the application, in order to achieve clear operation logic, a complete specified operation can be divided into several sub-operations. For example, the test of measuring the pH value of a solution can be composed of two sub-operations, which are liquid extraction operation and pH value reading colorimetry. Operation, wherein, the operation of taking liquid refers to the operation of dipping the liquid to be tested with a glass rod, and dropping the liquid to be tested on the PH test paper placed on the drip plate, and the colorimetric operation of pH value reading refers to the operation of using a colorimetric card and The color-changing PH test paper is used for colorimetric operation. Further, according to actual needs, one or more sub-operations may be selected from several sub-operations included in the complete specified operation as the operation to be assessed. The reading colorimetric operation is regarded as the operation to be assessed.

Based on this, in this step 202, the video containing the target object performing each operation to be assessed (hereinafter referred to as the target video) can be detected from the video stream collected by the camera when the target object performs the specified operation. As for how to detect the target video containing the target object performing each operation to be assessed from the video stream collected by the camera when the target object performs the specified operation, it will be described in detail below, and will not be described here.

Step 202: Determine the assessment score for the target object to perform the operation to be assessed according to the target video.

In the application, different target objects have different operating habits. For example, in the application scenario shown in Figure 1, some candidates are used to holding the glass rod in the right hand and the container in the left hand, while some candidates are used to holding the glass rod in the left hand. , hold the container in the right hand, as for whether to hold the glass rod in the left hand or the right hand, it will not affect the results of the chemical experiment operation. Instead, it is possible to set some specific assessment points for the operations to be assessed in the chemical experiment operation, and it is required that for different candidates, the operation of the specific assessment points should be standardized and unified.

For example, in the operation of taking liquid to be assessed, the specific assessment points include: the pH test paper is placed on the drip plate, the examinee uses a glass rod to dip the liquid to be tested and drops it on the pH test paper, and the PH test paper is placed in the dripping process. The test strips remain on the drip plate.

For another example, in the operation to be assessed, the pH value reading color comparison operation, the specific assessment points include: using a color comparison card, and bringing the color comparison card and the discolored PH test paper close to the color comparison.

Based on this, in this step 202, during the process of detecting the target object performing the operation to be assessed according to the target video, for the execution status of each assessment point of the operation to be assessed, determine the execution status of the target object to be assessed according to the execution status of each assessment point The assessment score for the operation. As for how it is implemented, there will be a detailed description below, which will not be described here.

As an embodiment, a highest assessment score can be set in advance for the operation to be assessed, and a corresponding deduction rule can be set for each assessment point of the operation to be assessed, for example, as shown in Table 1 below, for the liquid extraction operation and Examples of assessment rules for each colorimetric operation of pH readings:

Table 1

Based on the above Table 1 as an example, suppose that when the examinee performs the operation of taking liquid, he first places the pH test paper on the drip plate, and then pours the liquid to be tested on the pH test paper placed on the drip plate. The operation of using a glass rod to dip the liquid to be tested and drop it on the PH test paper is not standardized. Therefore, 20 points can be deducted from the highest assessment score of 40 points corresponding to the liquid extraction operation, and the candidate's assessment score for the liquid extraction operation is 20. point.

Step 203: Determine the assessment score of the target object performing the specified operation according to the assessment score of the target object performing the operation to be assessed.

It can be seen from the above description that the specified operation may include one or several operations to be assessed, and by performing step 202, the assessment score of each operation to be assessed by the target object can be determined, and accordingly, the assessment score of the operation to be assessed by the target object can be determined Determine the assessment score for the target object to perform the specified action.

As an example, the assessment scores of each operation to be assessed performed by the target object may be added to obtain the assessment score of the target object performing the specified operation.

As another example, corresponding score weights can be set for each operation to be assessed in advance, and based on the score weights, the assessment scores of the target object performing each operation to be assessed can be weighted and added to obtain the assessment of the target object performing the specified operation Fraction.

So far, the relevant description of the process shown in FIG. 2 is completed.

It can be seen from the above embodiments that by detecting the target video from the video stream collected by the camera when the target object performs a specified operation, the assessment score of the target object performing the operation to be assessed is determined according to the target video, and the assessment score of the target object performing the operation to be assessed is determined according to the score of the target object performing the operation to be assessed. The assessment score determines the assessment score for the target object to perform the specified operation, which realizes the automatic assessment of the process of the target object performing the specified operation. At the same time, because the uniform assessment standard is adopted for different target objects, it can effectively ensure the assessment of different target objects Objectivity and accuracy of results.

The following describes how to detect the target video containing the target object performing the operation to be assessed from the video stream collected by the camera when the target object performs the specified operation:

As an embodiment, the above-mentioned video streams can be identified in sequence, when an event (hereinafter referred to as the first video frame) for instructing the target object to start performing the operation to be assessed is identified from a video frame of the video stream (hereinafter referred to as the first video frame) An event), that is, when it is recognized that the first event exists in the first video frame of the video stream, it can be determined that the target object starts to perform the operation to be assessed. Similarly, when an event (hereinafter referred to as the second event) for instructing the target object to end the operation to be assessed is identified from a video frame after the first video frame (hereinafter referred to as the second video frame), the second video frame is identified. When the second event exists in the video frame, it is determined that the target object finishes performing the operation to be assessed. In the video stream, a segment of video in which the first video frame is the first video frame and the second video frame is the last video frame is determined as the target video corresponding to the operation to be assessed.

Taking the liquid taking operation as an example, the above-mentioned first event may refer to: unused pH test paper is placed on the drip plate, and the above-mentioned second event may refer to: the pH test paper placed on the drip plate changes color.

Through the above embodiments, it is realized that the target video including the target object performing the operation to be assessed is detected from the video stream collected by the camera when the target object performs the specified operation.

Please refer to FIG. 3 , which is an implementation process of step 202 provided by an exemplary embodiment of the present application.

As shown in Figure 3, the following steps are included:

Step 301: Input the target video into the trained deep learning model to obtain assessment information corresponding to the operation to be assessed.

Step 302: Determine the assessment score of the operation to be assessed according to the assessment information.

Step 302 and step 302 are described uniformly as follows:

First of all, the above-mentioned trained deep learning model can include two parts, namely the target detection network and the key point detection network. Among them, the target detection network can be implemented based on the YOLOV2 target detection method, and is configured to detect various equipment included in the input video frame, such as glass rods, drip plates, etc.; the key point detection network can be implemented based on the key point detection method, which is It is configured to locate the key area on the equipment included in the input video frame, for example, for the pH test paper, the key area is the discoloration area, and for example, for the glass rod, the key area is the head position.

It should be noted that in the embodiment of this application, the reason why the deep learning model is set to include two parts, the target detection network and the key point detection network, is that the target detection network only uses a rectangular frame to frame the equipment in the video frame. That is to locate the equipment in the video frame, and in some application scenarios, such as during the liquid extraction process, the head of the glass rod needs to be in contact with the pH test paper, and in order to detect this, it is necessary to use the key point detection network to detect the glass rod. Position the head of the rod for positioning.

In addition, those skilled in the art can understand that, in the application, video streams when multiple times of performing specified operations can be collected as training data, and each video frame in the training data is manually marked in advance, and the equipment, and then use the marked training data to train the deep learning model to obtain the above-mentioned trained deep learning model. As for the detailed training process of the deep learning model, this application will not describe it in detail.

Based on the above description, as an embodiment, the target video can be input into the trained deep learning model to obtain the assessment information corresponding to the operation to be assessed, and the assessment score of the operation to be assessed can be determined according to the assessment information.

Taking the application scenario of candidates performing chemical experiments as an example, by inputting the target video into the trained deep learning model, the position of the equipment on each video frame and the position of key areas on the equipment can be detected. Operate the corresponding assessment information. Since the actual operation process of the examinee can be determined through the position of the equipment on each video frame and the position of the key area on the equipment, based on this information, it is possible to determine the execution of each assessment point of the operation to be assessed by the examinee, so as to achieve The execution status of the assessment point determines the assessment score for the target object to perform the operation to be assessed.

In addition, as an example, before the target video is input into the trained deep learning model, preprocessing operations can also be performed on each video frame in the target video, where the preprocessing operations include but are not limited to the selection of the region of interest, image Resizing, image contrast adjustment, etc. Through this processing, the efficiency of recognizing the target video using the deep learning model and the accuracy of the recognition result can be improved.

In addition, for some specified operations, the target object can obtain an operation result information based on the actual operation result, for example, after the examinee performs the experiment of measuring the pH value of the solution, the measured pH value of the solution can be written on the answer sheet , the PH value is the operation result information obtained by the examinee.

Based on this, in this embodiment of the application, it is also possible to determine the assessment score of the operation result information obtained by the target object after performing the specified operation. The review score for the review operation determines the review score for the target object to perform the specified operation.

As an example, a highest assessment score can be set for the operation result information in advance, and a standard operation result information can be determined, and the assessment score of the operation result information obtained by the target object can be determined according to the highest assessment score and the standard operation result information.

Taking the experiment of measuring the pH value of the solution as an example, the flow shown in Figure 4 is shown below to illustrate the process of determining the assessment score of the operation result information obtained by the target object after performing the specified operation:

As shown in Figure 4, the following steps are included:

Step 401: Detect the discolored area on the pH test paper and the pH value of the solution measured by the examinee from the video stream.

First, how to detect the discolored area on the PH test paper from the video stream:

As an embodiment, the pH value to be read may be firstly detected from the video stream, that is, the video frame of the discolored area on the pH test paper to be detected (hereinafter referred to as the target video frame). After the examinee drops the solution to be tested on the PH test paper, the discoloration area will appear on the PH test paper, and the discoloration of the PH test paper is a process, and the timing of reading the pH value displayed by the PH test paper will affect the final reading. As a result, therefore, the video frame collected by the camera after the examinee drops the solution to be tested on the pH test paper can be determined as the target video frame.

In one example, for the pH value reading colorimetric operation, the video frame located after the second video frame in the video stream may be determined as the target video frame.

As an example, for each target video frame, the target video frame can be input to the above-mentioned deep learning model, so that the target detection network in the deep learning model can identify the PH test paper included in the target video frame and input the recognition result to The key point detection network in the deep learning model is used to locate the discoloration area in the target video frame including the PH test paper by the key point detection network according to the recognition result.

As another example, for each target video frame, traditional image processing techniques, such as binarization processing, foreground extraction, etc., can be used to locate the discolored area on the PH test paper included in the target video frame. The process will not be described in detail.

Next, explain how to detect the pH value of the solution measured by the candidate from the video stream:

As an embodiment, it is possible to detect from the video stream that there is a video frame of the pH value of the solution measured by the examinee, and then analyze and identify the video frame, for example, by using ORC (Optical Character Recognition, optical character recognition) technology, from the video frame Obtain the pH value of the solution measured by the examinee.

Step 402: Use the color information of the discolored area as a key to find the corresponding pH value in the color comparison card.

As an example, the above color information may be RGB color information, and may also be YUV color information, which is not limited in this application.

In this step 402, the color information of the discoloration area can be used as a key to search for a color comparison card, which includes the correspondence between color information and pH value, so as to find the corresponding pH value. The corresponding pH value is the standard operation result information described above.

Step 403: Determine the examination score according to the found pH value and the pH value of the solution measured by the examinee.

As an example, it can be known from the above description that by performing step 401 and step 402, multiple pH values, that is, multiple standard operation result information can be found. Based on this, in this step 403, as an example, for the found pH value corresponding to each target video frame, the difference between the pH value and the pH value of the solution measured by the examinee can be calculated, and then according to the calculated All differences are determined to determine the assessment score.

In one example, the average value of all differences can be calculated, and then the average value is compared with a plurality of preset ranges to determine which preset range the average value belongs to, and finally the preset range to which the average value belongs The corresponding assessment score is determined as the assessment score of the pH value of the solution measured by the examinee after performing the test of measuring the pH value of the solution.

So far, the relevant description of the process shown in FIG. 4 is completed.

Through the process shown in Figure 4, the automatic determination of the assessment score of the solution pH value measured by the examinee is finally realized.

Corresponding to the foregoing embodiments of the target detection method, the present application also provides embodiments of a target detection device.

Please refer to FIG. 5 , which is a block diagram of an embodiment of a target detection device provided in an exemplary embodiment of the present application. The device includes:

The detection module 51 is configured to detect the target video from the video stream collected by the camera when the target object performs a specified operation, and the target video includes the target object to be assessed in the process of performing the specified operation operation video;

The first assessment module 52 is configured to determine the assessment score for the target object to perform the operation to be assessed according to the target video;

The second assessment module 53 is configured to determine the assessment score for the target object to perform the specified operation according to the assessment score for the target object to perform the operation to be assessed.

In one embodiment, the detection module 51 detects the target video from the video stream collected by the camera when the target object performs a specified operation, including:

Identifying the video frames in the video stream in sequence, and determining the target when it is identified that there is a first event in the first video frame of the video stream indicating that the target object starts to perform the operation to be assessed The object starts to perform the operation to be assessed, and when it is recognized that a second event indicating that the target object ends the operation to be assessed exists in a second video frame after the first video frame, determining the target object End the execution of the operation to be assessed;

Determining a section of video in which the first video frame is the first video frame and the second video frame is the last video frame in the video stream as the target video corresponding to the operation to be assessed.

In one embodiment, the first assessment module 52 determines the assessment score of the target object performing the operation to be assessed according to the target video, including:

Inputting the target video into a trained deep learning model to obtain assessment information corresponding to the operation to be assessed;

An assessment score of the operation to be assessed is determined according to the assessment information.

In one embodiment, the device further includes (not shown in FIG. 5 ):

The third evaluation module is used to determine the evaluation score of the operation result information obtained by the target object after performing the specified operation;

The second assessment module 53 determining the assessment score of the target object performing the specified operation according to the assessment score of the target object performing the operation to be assessed includes:

The assessment score of the target object performing the specified operation is determined according to the assessment score of the operation result information and the assessment score of the target object performing the operation to be assessed.

In one embodiment, the specified operation performed by the target object is: the examinee performs the experiment of measuring the pH value of the solution;

The third evaluation module determines the evaluation score of the operation result information obtained by the target object after performing the specified operation includes:

Detect the discoloration area on the pH test paper and the pH value of the solution measured by the examinee from the video stream;

Using the color information of the discolored area as a key to find the corresponding pH value in the color comparison card;

The assessment score is determined based on the found PH value and the solution PH value measured by the examinee.

In one embodiment, the specified operation performed by the target object is: the examinee performs the experiment of measuring the pH value of the solution;

The operation to be assessed is: the examinee performs the liquid taking operation and the pH value reading colorimetric operation in the experimental process of measuring the pH value of the solution;

The liquid-taking operation refers to the operation of dipping the liquid to be tested with a glass rod, and dropping the liquid to be tested on the pH test paper placed on the drip plate;

The pH value reading operation refers to the operation of using a color comparison card and a color-changing pH test paper to perform color comparison.

Please continue to refer to FIG. 6 , the present application further provides an electronic device, including a processor 601 , a communication interface 602 , a memory 603 , and a communication bus 604 .

Wherein, the processor 601, the communication interface 602, and the memory 603 communicate with each other through the communication bus 604;

Memory 603, used to store computer programs;

The processor 601 is configured to execute the computer program stored in the memory 603. When the processor 501 executes the computer program, the steps of the object detection method provided in the embodiment of the present application are realized.

The present application also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the object detection method provided in the embodiment of the present application are implemented.

For the implementation process of the functions and effects of each unit in the above device, please refer to the implementation process of the corresponding steps in the above method for details, and will not be repeated here.

As for the device embodiment, since it basically corresponds to the method embodiment, for related parts, please refer to the part description of the method embodiment. The device embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed to multiple network elements. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this application. It can be understood and implemented by those skilled in the art without creative effort.

Other embodiments of the present application will be readily apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any modification, use or adaptation of the application, these modifications, uses or adaptations follow the general principles of the application and include common knowledge or conventional technical means in the technical field not disclosed in the application . The specification and examples are to be considered exemplary only, with a true scope and spirit of the application indicated by the following claims.

It should also be noted that the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus comprising a set of elements includes not only those elements, but also includes Other elements not expressly listed, or elements inherent in the process, method, commodity, or apparatus are also included. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising said element.

The above is only a preferred embodiment of the application, and is not intended to limit the application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the application should be included in the application. within the scope of protection.

Claims (8)

1. A method of target detection, the method comprising:

detecting a target video from a video stream acquired by a camera when a target object executes a specified operation, wherein the target video is a video containing an operation to be checked executed by the target object in the process of executing the specified operation;

inputting the target video into a trained deep learning model to obtain assessment information corresponding to the operation to be examined; determining the assessment score of the operation to be assessed according to the assessment information;

the method comprises the steps of setting the highest assessment score for operation result information in advance, determining standard operation result information, and determining the assessment score of the operation result information obtained by a target object after the specified operation is executed according to the highest assessment score and the standard operation result information;

and determining the assessment score of the target object for executing the specified operation according to the assessment score of the operation result information and the assessment score of the target object for executing the operation to be assessed.

2. The method of claim 1, wherein detecting the target video from the video stream acquired by the camera when the target object performs the specified operation, comprises:

identifying video frames in the video stream in sequence, determining that the target object starts to execute the operation to be checked when a first event for indicating the target object to start executing the operation to be checked exists in a first video frame of the video stream, and determining that the target object ends to execute the operation to be checked when a second event for indicating the target object to end the operation to be checked exists in a second video frame after the first video frame is identified;

and determining a section of video taking the first video frame as a first video frame and the second video frame as a last video frame in the video stream as a target video corresponding to the operation to be checked.

3. The method according to claim 1, characterized in that:

the target object performs a specified operation as: the examinee performs an experiment for measuring the pH value of the solution;

the determining the assessment score of the operation result information obtained by the target object after the specified operation is executed comprises the following steps:

detecting a color-changing area on the PH test paper and the PH value of the solution measured by the examinee from the video stream;

the color information of the color change area is used as a keyword to find the corresponding PH value in the color chart;

and determining the assessment score according to the searched pH value and the pH value of the solution measured by the examinee.

4. The method according to claim 1 or 2, characterized in that:

the target object performs a specified operation as: the examinee performs an experiment for measuring the pH value of the solution;

the operation to be checked is as follows: the examinee executes liquid taking operation and PH value reading colorimetric operation in the experiment process of measuring the PH value of the solution;

the liquid taking operation is to dip the liquid to be measured by using a glass rod and drop the liquid to be measured on PH test paper placed on a drop plate;

the PH value reading colorimetric operation refers to an operation of using a colorimetric card to conduct colorimetric with color-changing PH test paper.

5. An object detection device, the device comprising:

the detection module is used for detecting a target video from a video stream acquired by a camera when a target object executes a specified operation, wherein the target video is a video containing an operation to be checked executed by the target object in the process of executing the specified operation;

the first assessment module is used for inputting the target video into a trained deep learning model to obtain assessment information corresponding to the operation to be examined; determining the assessment score of the operation to be assessed according to the assessment information;

the second assessment module is used for setting the highest assessment score for the operation result information in advance, determining standard operation result information and determining the assessment score of the operation result information obtained by the target object after the specified operation is performed according to the highest assessment score and the standard operation result information; and determining the assessment score of the target object for executing the specified operation according to the assessment score of the operation result information and the assessment score of the target object for executing the operation to be assessed.

6. The apparatus of claim 5, wherein the detection module detects the target video from a video stream acquired by the camera when the target object performs the specified operation, comprising:

identifying video frames in the video stream in sequence, determining that the target object starts to execute the operation to be checked when a first event for indicating the target object to start executing the operation to be checked exists in a first video frame of the video stream, and determining that the target object ends to execute the operation to be checked when a second event for indicating the target object to end the operation to be checked exists in a second video frame after the first video frame is identified;

and determining a section of video taking the first video frame as a first video frame and the second video frame as a last video frame in the video stream as a target video corresponding to the operation to be checked.

7. An electronic device comprising a readable storage medium and a processor;

wherein the readable storage medium is for storing machine executable instructions;

the processor is configured to read the machine-executable instructions on the readable storage medium and execute the instructions to implement the steps of the method of any of claims 1-4.

8. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program which, when executed by a processor, implements the steps of the method of any of claims 1-4.